Authors:
Ana González-Marcos
1
;
Rubén Olarte-Valentín
1
;
Joaquín Ordieres-Meré
2
and
Fernando Alba-Elías
1
Affiliations:
1
Department of Mechanical Engineering, Universidad de La Rioja, c/ San José de Calasanz 31, 26004 Logroño, La Rioja and Spain
;
2
PMQ Research Group, ETSII, Universidad Politécnica de Madrid, José Gutiérrrez Abascal 2, 28006 Madrid and Spain
Keyword(s):
Educational Data Mining, Students’ Performance, Project Management, Higher Education.
Abstract:
This work presents a predictive analysis of the academic performance of students enrolled in project management courses in two different engineering degree programs. Data were gathered from a virtual learning environment that was designed to support the specific needs of the proposed learning experience. The analyzed data included individual attributes related to communication, time, resources, information and documentation activity, as well as behavioral assessment. Also, students’ marks on two exams that took place during the first half of the course were considered as input variables of the predictive models. Results obtained using several regression and classification algorithms –support vector machines, random forests, and gradient boosted trees– confirm the usefulness of Educational Data Mining to predict students’ performance. These models can be used for early identification of weak students who will be at risk in order to take early actions to prevent these students from fai
lure.
(More)